Apache AGE (incubating) for openGauss ​

Before You Start ​

Apache AGE is used by creating an extension.

create extension age;

After AGE is installed, the ag_catalog schema is created by default. All built-in data types and functions of AGE are stored under ag_catalog. Therefore, when using AGE, especially when executing Cypher statements, you need to first run the following command:

SET search_path TO ag_catalog;

You also need to run the load 'age' command to ensure that all hooks of the AGE extension are loaded for graph data integrity.

load 'age';

1. Graph Operations ​

A graph consists of a set of vertices and edges, where each vertex and edge has a property map. A vertex is the basic object of a graph, which can exist independently of anything else in the graph. An edge creates a directed connection between two vertices.

1.1 Creating a Graph ​

To create a graph, use create_graph(graph_name):

SELECT * FROM ag_catalog.create_graph('graph_name');

1.2 Deleting a Graph ​

To delete a graph, use drop_graph(graph_name, cascade).

1. The first parameter graph_name is the graph to be deleted, and the second parameter cascade is a boolean value indicating whether to delete graph label and data.
2. It is recommended to set cascade to true; otherwise, all contents in the graph must be manually deleted using SQL DDL commands

SELECT * FROM ag_catalog.drop_graph('graph_name', true);

1.3 Creating Vertex Labels in a Graph ​

To create a vertex label in a graph, use create_vlabel(graph_name, label_name).

SELECT * FROM ag_catalog.create_vlabel('graph_name','label_name');

Whenever a vertex label is created using the create_vlabel() function, a corresponding table named <label_name> is generated within the namespace new_graph. The same behavior applies to create_elabel() for creating edge labels. These two functions are not mandatory. if a label table does not already exist when creating vertices or edges via Cypher, it will be created automatically.

1.4 Creating Edge Labels in a Graph ​

To create a graph, use create_elabel(graph_name, label_name).

SELECT * FROM ag_catalog.create_elabel('graph_name','label_name');

2. Graph Storage ​

2.1 Graph ​

After a graph is created, the kernel creates a schema with the same name as the graph. Meanwhile, a record is inserted into the ag_catalog.ag_graph table, marking the newly created schema aas the storage location for the graph data.

SELECT create_graph('new_graph');

NOTICE:  graph "new_graph" has been created
 create_graph 
--------------

(1 row)

SELECT * FROM ag_catalog.ag_graph;

   name    | namespace 
-----------+-----------
 new_graph | new_graph
(1 row)

Vertices and Edges in a Graph ​

-- After a graph is created, two tables, _ag_label_vertex and _ag_label_edge, are created in the schema corresponding to the graph as the default vertex table and edge table. Meanwhile, two records are inserted into ag_catalog.ag_label to associate the vertex table and edge table with the schema.
SELECT * FROM ag_catalog.ag_label;

       name       | graph | id | kind |          relation          
------------------+-------+----+------+----------------------------
 _ag_label_vertex | 68484 |  1 | v    | new_graph._ag_label_vertex 
 _ag_label_edge   | 68484 |  2 | e    | new_graph._ag_label_edge   
(2 rows)

-- Create a vertex table.
SELECT create_vlabel('new_graph', 'Person');
NOTICE:  VLabel "Person" has been created
 create_vlabel 
---------------
 
(1 row)

-- After a vertex table is created, a record is inserted into the ag_catalog.ag_label table to associate the vertex table with the graph. The kind 'v' represents a vertex table and the kind 'e' represents an edge table.
SELECT * FROM ag_catalog.ag_label;
       name       | graph | id | kind |          relation          
------------------+-------+----+------+----------------------------
 _ag_label_vertex | 68484 |  1 | v    | new_graph._ag_label_vertex 
 _ag_label_edge   | 68484 |  2 | e    | new_graph._ag_label_edge   
 Person           | 68484 |  3 | v    | new_graph."Person"         
(3 rows)

3. Cypher Query ​

Cypher statements cannot be executed directly in the database.Instead, they must be passed as arguments to the cypher() function, which then returns a SETOF records.

3.1 Introduction to cypher() ​

cypher(graph_name, query_string, parameters)

graph_name is the graph to query. query_string is the Cypher statement. parameters is optional and can only be used with Prepared Statements; otherwise, an error will be thrown.

SELECT * FROM cypher('graph_name', $$ 
/* Cypher Query Here */ 
$$) AS (result1 agtype, result2 agtype);

Note

1. (result1 agtype, result2 agtype) after AS, i.e., SETOF records, must match the number of return values in the Cypher statement.
2. SELECT * FROM cypher cannot be written as SELECT cypher.
3. Before executing the statement, execute load 'age' and set search_path = ag_catalog.

Data Types and Functions ​

The following provides a brief description of the data types, Cypher statements, and functions included in AGE.

1 Data Types ​

AGE creates two data types: graphid and agtype. graphid represents the unique ID identifier for vertices and edges. agtype is the core data type of AGE. For more details, refer to https://age.apache.org/age-manual/master/intro/types.html.

1.1 Simple Data Types ​

Simple data types include Null, Integer, Float, Numeric, Bool, and String.

1.2 Composite Data Types ​

Composite data types include List and Map.

• List operations

• Map operations

1.3 Simple Entities ​

Simple entity types include GraphId, Labels, and Properties, and simple entity types can further compose Vertex, Edge, and Composite Entities.

1.4 Vertex ​

A Vertex is the fundamental building block of a graph, representing a node.

1.5 Edge ​

An Edge is the basic building block of a graph, representing an edge.

1.6 Composite Entities ​

A Path composed of Vertex and Edge.

2 Cypher Statements ​

For detailed information on AGE's support for Cypher statements, refer to https://age.apache.org/age-manual/master/clauses/match.html.

2.1 Match ​

2.2 WITH ​

2.3 SKIP ​

2.4 LIMIT ​

2.5 Return ​

2.6 ORDER BY ​

2.7 CREATE ​

2.8 SET ​

2.9 REMOVE ​

2.10 DELETE ​

3 Functions ​

Functions primarily involve operations on agtype and the generation of expressions. For details, refer to https://age.apache.org/age-manual/master/functions/predicate_functions.html.

3.1 Predicate Functions ​

3.2 Scalar Functions ​

3.3 List Functions ​

3.4 Numeric Functions ​

3.5 Logarithmic Functions ​

3.6 Trigonometric Functions ​

3.7 String Functions ​

3.8 Aggregation Functions ​